[16:10:32] <Ivoz> dstufft, 'nother nag for virtualenv
[16:10:59] <Ivoz> or qwcode you interested? want to put out release
[16:13:25] <qwcode> Ivoz, I let dstufft do that : ) usually we do pip and virtualenv at the same time... you want to do it separate?
[16:13:57] <Ivoz> qwcode, ya... it's broken for python 3.5
[16:14:24] <Ivoz> Current develop will fix non-windows-3.5
[16:15:06] <Ivoz> I've been working on trying to fix as many small niggles as possible people have had
[16:15:22] <qwcode> Ivoz, roger, how about send out email to pypa-dev and state the case
[16:15:46] <Ivoz> while the rewrite branch makes its merry way towards stable
[16:16:55] <Ivoz> Also wondering about our status with gpg-signing things....
[16:17:14] <Ivoz> ATM is the bus factor 1 for that, for each project?
[16:17:30] <qwcode> Ivo, yea, I was thinking about that.
[16:18:06] <Ivoz> Surely if we want it more than just "oh look a pgp signature" we'd want to get a PyPA signing key
[16:18:20] <Ivoz> to then trust one or two pypa devs for each project
[16:20:01] <Ivoz> Is there some sort of framework for keys saying they trust other keys to make releases for X project(s)?
[16:21:17] <Ivoz> haha sure enough donald wrote a blog on this... /reads
[16:36:52] <Ivoz> I also noticed we don't have any packaging history for 2015 :D https://www.pypa.io/en/latest/history/
[16:36:56] <Ivoz> Surely we've done some stuff this year
[16:44:00] <Ivoz> Hmm, I think the best point we could get involved package signing was at least saying "This person created X project"
[16:45:41] <Ivoz> or, in a more complicated case, X key is trusted for B package, and X key trusts Y and Z keys to release packages for B, and this release of the package was signed by Z, so it should be trustworthy"
[18:17:05] <Ivoz> qwcode, was thinking about asking for that to go in develop instead
[18:17:29] <qwcode> Ivoz, yea, I didn't notice : (
[18:17:55] <Ivoz> If you merge master into develop now, will git complain about anything in the future?
[18:18:29] <Ivoz> Like when we pull back develop into master to cut a new release
[18:21:39] <qwcode> it probably be ok, would have to look at the merge, but I'd prefer on principle not to do it in reverse like that.... shrug
[18:24:13] <Ivo> yay accidentally running two irc clients
[20:28:10] <chogan> Hi. I'm trying to build setuptools with a python I built with the Intel compiler on OS X by running python bootstrap.py, but for some reason it's unable to import setuptools.command in setuptools/command/__init__.py. Any advice?
[20:29:50] <chogan> Here's the full stack trace: Regenerating egg_info
[20:38:44] <chogan> Yeah, I tracked it to that import statement with the debugger
[20:39:48] <Ivo> chogan, if you don't care about how "pure" your bootstrap is, you can also use a zip of pip to install a setuptools wheel
[20:40:00] <qwcode> chogan, Ivo, see here https://pip.pypa.io/en/stable/installing/#install-pip one of the examples explains using options with get-pip.py to install from local distributions
[20:42:03] <Ivo> chogan, get-pip.py, at it's barest mechanics, is just doing python -m <pip.whl renamed to be a zip> install <pip.whl> <setuptools.whl>
[20:42:51] <Ivo> chogan, also, have you tried python -m ensurepip ??
[20:43:25] <chogan> Ivo: That gives the same segmentation fault.